ORIGINAL RESEARCHpublished: 08 June 2021

doi: 10.3389/fvets.2021.676499

Frontiers in Veterinary Science | www.frontiersin.org 1 June 2021 | Volume 8 | Article 676499

Edited by:

Stephen Raverty,

Animal Health Center, Canada

Reviewed by:

Weigang Xu,

Naval Medical University, China

Sophie Dennison,

TeleVet Imaging Solutions, PLLC,

United States

Claudia Kusmic,

Italian National Research Council, Italy

*Correspondence:

Maria José Caballero

mariajose.caballero@ulpgc.es

Marisa Andrada

marisaana.andrada@ulpgc.es

Specialty section:

This article was submitted to

Veterinary Experimental and

Diagnostic Pathology,

a section of the journal

Frontiers in Veterinary Science

Received: 05 March 2021

Accepted: 26 April 2021

Published: 08 June 2021

Citation:

Velázquez-Wallraf A, Fernández A,

Caballero MJ, Møllerløkken A,

Jepson PD, Andrada M and Bernaldo

de Quirós Y (2021) Decompressive

Pathology in Cetaceans Based on an

Experimental Pathological Model.

Front. Vet. Sci. 8:676499.

doi: 10.3389/fvets.2021.676499

Decompressive Pathology inCetaceans Based on an ExperimentalPathological Model

Alicia Velázquez-Wallraf 1, Antonio Fernández 1, Maria José Caballero 1*,

Andreas Møllerløkken 2, Paul D. Jepson 3, Marisa Andrada 1* and Yara Bernaldo de Quirós 1

1 Veterinary Histology and Pathology, Atlantic Center for Cetacean Research, University Institute of Animal Health and Food

Safety (IUSA), Veterinary School, University of Las Palmas de Gran Canaria, Canary Islands, Spain, 2 Faculty of Engineering,

Norwegian University of Science and Technology, NTNU, Trondheim, Norway, 3 Institute of Zoology, Zoological Society of

London, London, United Kingdom

Decompression sickness (DCS) is a widely known clinical syndrome in human medicine,

mainly in divers, related to the formation of intravascular and extravascular gas bubbles.

Gas embolism and decompression-like sickness have also been described in wild

animals, such as cetaceans. It was hypothesized that adaptations to the marine

environment protected them from DCS, but in 2003, decompression-like sickness was

described for the first time in beaked whales, challenging this dogma. Since then, several

episodes of mass strandings of beaked whales coincidental in time and space with naval

maneuvers have been recorded and diagnosed with DCS. The diagnosis of human DCS

is based on the presence of clinical symptoms and the detection of gas embolism by

ultrasound, but in cetaceans, the diagnosis is limited to forensic investigations. For this

reason, it is necessary to resort to experimental animal models to support the pathological

diagnosis of DCS in cetaceans. The objective of this study is to validate the pathological

results of cetaceans through an experimental rabbit model wherein a complete and

detailed histopathological analysis was performed. Gross and histopathological results

were very similar in the experimental animal model compared to stranded cetaceans with

DCS, with the presence of gas embolism systemically distributed as well as emphysema

and hemorrhages as primary lesions in different organs. The experimental data reinforces

the pathological findings found in cetaceans with DCS as well as the hypothesis that

individuality plays an essential role in DCS, as it has previously been proposed in animal

models and human diving medicine.

Keywords: gas bubble, stranded cetaceans, pathology, rabbit model, decompression sickness

INTRODUCTION

Decompression sickness (DCS) is a widely known clinical syndrome in humanmedicine, mainly inrecreational and professional divers. It is considered to occur when the sum of the gases dissolvedin the tissues exceeds the environmental pressure, causing the formation of intravascular andextravascular gas bubbles. The presence of gas embolism as noted by ultrasound must be observedfor its confirmatory clinical diagnosis. In these cases, the patient is treated with a hyperbaricchamber. If the gas and the symptoms resolve, the diagnosis of DCS is definitive (1). In forensic

Velázquez-Wallraf et al. Decompressive Pathology Through Experimental Model

investigations, gas bubble-related lesions are the main findings(2, 3). These bubbles can result in mechanical, biochemical, andembolic damage with different severity levels depending on theirnumber and their size (1). The respiratory system is the mostaffected organ by DCS when the amount of bubbles exceeds thepulmonary capillaries’ capacity to eliminate them, resulting insevere lung damage (4).

Cetaceans are mammals that returned to the marineenvironment 60million years ago and have developed behavioral,anatomical, and physiological adaptations for this new habitat,including those related to diving (5). It was hypothesized thatthese adaptations protected them from possible DCS, but in2003, lesions compatible with DCS were described for the firsttime in beaked whales stranded coincidentally in time and spacewith naval exercises using high-intensity and mid-frequencyactive sonars (6, 7). This first description of a decompression-like sickness in beaked whales broke the dogma that cetaceanswere immune to this disease (7). These findings have also beenfound in other beaked whale (BW) strandings associated withnaval maneuvers (8–11) as well as in Risso’s dolphins but, inthis case, were caused naturally due to an interaction with theirpreys during feeding (12). Gas embolism and decompression-likesickness have also been described in other wild animals such assea turtles (13).

In cetaceans, the diagnosis of DCS is limited to forensicinvestigations and its pathological gas bubble-associated lesionssince a clinical diagnosis is not possible due to obvious logisticaland ethical restrictions. For this reason, it is necessary to resortto experimental animal models to contrast the macroscopicand microscopic lesions and to support the pathologicaldiagnosis described in marine mammals affected by DCS. Toour knowledge, there are very few publications focused onthe pathological study of DCS in humans or other species(14–17), and almost all the articles on animal experimentationfocus on the analysis of specific tissues for the applicationof preventive treatments in DCS (4, 18–20). Furthermore,there is no pathological comparative study showing gross andhistological findings in experimental and natural DCS. Therefore,the objective of this study is to validate the pathological resultsof cetaceans through an experimental model wherein a completeand detailed histopathological analysis is performed. For thispurpose, an experimental rabbit model was performed in whichsevere DCS is reproduced, presenting the pathological results inthese animals and then comparing them with the pathologicalfindings in cetaceans.

MATERIALS AND METHODS

For this study, 18 males of New Zealand white rabbits of3.15 ± 0.65 kg were used. These animals were divided intocompression/decompression model (C/D) (n = 14) and controlgroup (C) (n= 4).

All experiments were accomplished following the EuropeanUnion’s laboratory animals’ regulation and were conductedunder surgical anesthesia: subcutaneous injections ofmedetomidine (0.5 mg/kg) and ketamine (25 mg/kg). The

C/D model was carried out in the experimental animalfacilities of St. Olav University Hospital NTNU, Norway(Trondheim, Norway), and the Norwegian Committee forAnimal Experiments approved the protocol (2154). The controlgroup was carried out in the experimental animal facilities ofDr. Negrín University Hospital (Las Palmas de Gran Canaria,Spain), and the Ethical Committee for Animal Experiments ofthe University of Las Palmas de Gran Canaria approved theprotocol (CEEBA-HUGCDN 002/2010).

The Compression/Decompression ModelThe rabbits were anesthetized with the protocol described aboveand compressed in pairs in a dry, hyperbaric chamber (AnimalChamber System, NUT, Haugesund, Norway) with a divingprofile selected to induce severe decompression stress withexcessive amounts of intra-corporal gas formation: eight absoluteatmospheres during 45min, followed by fast decompression(0.33 m/s) to one atmosphere (21). One animal appeared deadwhen recovered from the chamber, and it remains unclear at whattime during the treatment the animal died; thus, it was withdrawnfrom the study. The animals were monitored for 1 h afterdecompression. A group of animals (n = 8) died within 25minpost-decompression (C/Dmortality group), while the rest (n= 5)survived the observation period of 1 h and were euthanized withan intraperitoneal injection of diluted pentobarbital (200 mg/kg)(C/D euthanized group).

Control GroupThe rabbits were first anesthetized with the protocoldescribed above and later euthanized with an intraperitonealinjection of diluted pentobarbital (200 mg/kg) as in the C/Deuthanized group.

Pathological StudyNecropsy was carried out for each rabbit in a dorsal decubitusposition. Dissection was carefully done to avoid severing largeblood vessels following the method of Bernaldo de Quiróset al. (22) to characterize the presence of intravascular andextravascular gas bubbles using a gas score index. This index-basedmethod consists of giving a gas score from 0 to 6 for each ofthe defined vascular locations (i.e., subcutaneous veins, femoralveins, mesenteric veins, caudal vena cava, coronary veins, andto the right atrium) and a gas score from 0 to 3 to describe thepresence and distribution of extravascular gas (i.e., subcapsularand interstitial emphysema) that may affect different organs. Thesum of the gas score of each intravascular and extravascularlocation calculates the total gas score in each rabbit. In the currentstudy, the mode of each group for intra- and extravascularlocations has been calculated.

Representative samples of the lung, trachea, superficialcervical lymph node, spleen, central nervous system,heart, liver, stomach, small and large intestine, mesentericlymph node, kidney, urinary bladder, and skeletal muscle(gastrocnemius) were collected and fixed in 10% bufferedformalin. These tissues were processed routinely andembedded in paraffin wax, and 5-µm-thick sectionswere cut and stained with hematoxylin and eosin (23)

Frontiers in Veterinary Science | www.frontiersin.org 2 June 2021 | Volume 8 | Article 676499

Velázquez-Wallraf et al. Decompressive Pathology Through Experimental Model

FIGURE 1 | Gas score modes of C/D mortality animals, C/D euthanized animals, and control group in the different locations selected.

for microscopic analysis. Histological sections fromthe heart and skeletal muscle were also stained withphosphotungstic acid hematoxylin and Masson’s trichrome(23), respectively, to evidence changes in skeletal andcardiac musculature.

Comparison With CetaceanDecompressive PathologyThe histopathological results from the animal model werecompared with the necropsy reports and histology slides fromstranded cetaceans that have been studied and diagnosedwith DCS by our research group. This included 31 animals:eight Cuvier’s BWs (Ziphius cavirostris), one Blainville’s BW(Mesoplodon densirostris), and one Gervais’s BW (Mesoplodoneuropaeus) stranded in the islands of Fuerteventura andLanzarote (Spain) in 2002 (7); four Cuvier’s BWs stranded onthese same islands in 2004 (8); four Cuvier’s BWs stranded in2006 and one Cuvier’s BWs stranded in 2011, both in Almeria(Spain) (9, 10); 10 Cuvier’s BWmass stranding in Corfú (Greece)in 2011 (11), all of them coincidental in time and space withnaval exercises; and two Risso’s dolphins (Grampus griseus) thatwere diagnosed with a decompressive disease after an interactionwith a prey (12). Since the pathological results of these animalshave already been published, the comparison with the originalpathological results from this study will be addressed in theDiscussion section.

RESULTS

Presence, Distribution, and Amount ofBubblesRabbits from the control group presented very few or an absenceof gas bubbles. One animal presented few gas bubbles in themesenteric veins, occasional bubbles in the subcutaneous veins,and scarce bubbles in the adipose tissue (total gas score: 4).One more animal presented only occasional gas bubbles in themesenteric veins (total gas score: 1). These results were previouslyreported by Bernaldo de Quirós et al. (22). The resulting gas scoremode calculated in this study for all locations was 0 (Figure 1)since the two remaining animals showed no bubbles.

In rabbits from the C/D mortality group, gas bubbles wereobserved in abundant numbers and/or filling complete vesselsections in the subcutaneous veins, the femoral veins, themesenteric veins (Figure 2E), the caudal vena cava, the rightatrium (Figure 2C), and the coronary veins. In addition, a sparseor moderate presence of subcapsular and interstitial emphysemawas observed. The gas score mode in the subcutaneous veins,the femoral vein, and the caudal vena cava was 6, while themesenteric and the coronary veins, along with the right atrium,had a gas score mode of 5 (Figure 1). The gas score modefor interstitial emphysema in this group was 2, while that ofsubcapsular emphysema was 0. The total gas score ranged from29 to 40 (22). Additionally, large amounts of gas bubbles werefound disseminated through other vascular locations.

Frontiers in Veterinary Science | www.frontiersin.org 3 June 2021 | Volume 8 | Article 676499

Velázquez-Wallraf et al. Decompressive Pathology Through Experimental Model

Gas bubbles were not found in the C/D euthanized group, witha total gas score of 0 in all animals and a gas score mode for alllocations of 0 (22).

Gross Examination and HistopathologyControl GroupGross findings in this group showed congestion in differentorgans, such as lung (3/4, 75%), liver (3/4, 75%), kidney(3/4, 75%), spleen (2/4, 50%), and brain (1/4, 25%), andmild multifocal petechial hemorrhages in thymus (3/4,75%). These findings were confirmed histologically. Noother histopathological findings were observed, excepthypereosinophilia (2/4, 50%) and vacuolization (2/4, 50%)in muscular cardiac fibers and vacuolization of hepatocytes(1/4, 25%).

C/D Model: Mortality GroupEmphysema was the predominant lesion observed in the lungof C/D mortality animals (6/8, 75%), with grossly voluminous,pale, and gas-distended pulmonary areas (Figure 2A). Otherlung findings were congestion (3/8, 38%) and alveolar edemaevidenced by exudation of fluid from the cut surface (2/8,25%). The lung’s microscopic appearance showed mild to severeemphysema in all animals (8/8, 100%). Besides these, mildpulmonary congestion (7/8, 88%) and alveolar hemorrhagesranging from mild focal hemorrhages to severe multifocalhemorrhages (4/8, 50%) as well as microscopic intravascularbubble-like round empty spaces surrounded by blood cells (3/8,38%) were observed (Figure 3A).

Marked subcapsular splenic emphysema was observed (3/8,38%). Histological emphysema was confirmed in six out of eightanimals (75%). Gross cavities underneath the splenic capsule andwithin splenic parenchyma were observed microscopically alongwith mild splenic congestion (7/8, 88%). Cerebral congestionwas seen in four cases of C/D mortality animals (4/8, 50%),and two animals showed, microscopically, mild local to extensivehemorrhages in the meningeal area and around cerebralcapillaries (2/8, 25%) (Figure 3E).

The most relevant finding in the heart was the presence ofhemorrhages, mainly in the right ventricle, in the group of C/Dmortality animals (3/8, 38%). Mild hemorrhages were confirmedmicroscopically (4/8, 50%) (Figure 3C), while two showedbubble-like cavities in cardiac capillaries (2/8, 25%). Acutechanges such as mild hyaline changes of muscle fibers as wellas hypereosinophilia and mild intracytoplasmic vacuolization ofinjured cardiomyocytes were observed in seven animals out ofeight (88%). In contrast, contraction band fibers were detected inonly one animal (1/8, 13%). Cardiac muscle fibers were separatedby expanded interstitial spaces filled with pale pink material,indicating mild interstitial edema (3/8, 38%).

In the liver of C/D mortality animals, mild hepatomegalyand general congestion were the main gross findings (5/8,63%). Histologically, moderate congestion was confirmed by thedistention of central veins and sinusoids. Mild vacuolization ofcentrilobular hepatocytes with cytoplasmic ballooning of thesecells was also observed (4/8, 50%). Renal congestion was also

FIGURE 2 | Macroscopic lesions compared between decompression

sickness in rabbits (left row) and cetaceans (right row). (A) Presence of pale

lungs with distended and enlarged areas, mainly denoted in the right side, in a

rabbit dead by C/D protocol. (B) Emphysematous lungs in a beaked whale

diagnosed with decompression-like sickness. (C) Heart of a C/D mortality

rabbit with congestion and macroscopic bubbles in the right atrium (white

arrow) and caudal vena cava (white star). (D) Heart of a beaked whale with

dilated right atrium due to the presence of macroscopic bubbles, which are

also observed in the coronary vessels (white arrows). (E) Mesenteric area of a

C/D mortality rabbit. Emphysematous fat and congestion of blood vessels

running through mesenteric fat. The presence of gas bubbles in the

mesenteric veins is denoted (white arrow). (F) Mesenteric area of a Risso’s

dolphin with visible bubbles circulating in the mesenteric vessels (white arrows)

and congestion.

found in this group (4/8, 50%). Microscopic vascular bubble-likecavities were also observed (2/8, 25%).

Other findings in this group were emphysema (7/8, 88%) andvascular congestion (4/8, 50%) associated with the adipose tissue.At the microscopic analysis of this group’s skeletal muscle, acutechanges such as mild hypereosinophilia were found in sevenanimals out of eight (88%) as well as interfibrillar mild interstitialedema (4/8, 50%).

C/D Model: Euthanized GroupThe lung of C/D euthanized animals showed emphysema(2/5, 40%), while edema was present in one animal of five(20%). Histologically, 100% of animals presented mild lungemphysema, mild congestion (3/5, 60%), and multifocal alveolarhemorrhages (1/5, 20%). While splenic subcapsular emphysemawas only macroscopically observed in one animal (1/5, 20%) andcongestion in two animals (2/5, 40%), the microscopic analysisrevealed mild splenic congestion in 100% of animals (5/5, 100%)and subcapsular and parenchymal emphysema in 80% of animals

Frontiers in Veterinary Science | www.frontiersin.org 4 June 2021 | Volume 8 | Article 676499

Velázquez-Wallraf et al. Decompressive Pathology Through Experimental Model

FIGURE 3 | Histological findings compared between decompression sickness

in rabbits (left row) and cetaceans (right row) stained routinely with

hematoxylin–eosin. (A) C/D mortality rabbit. Intravascular bubble-like round

empty space among blood cells (black star), mild emphysema, and congestion

in pulmonary areas. (B) Risso’s dolphin, ×10. Microscopic bubble-like cavities

circulating within a pulmonary blood vessel (black star). (C) C/D mortality

rabbit, ×20. Multifocal hemorrhages (black arrows) and congestion of cardiac

capillaries. (D) Risso’s Dolphin, ×10. Presence of congestive capillaries in

cardiac tissue (black arrows). (E) C/D mortality rabbit, ×20. Hemorrhages in

the subarachnoid area of the central nervous system. (F) Beaked whale, ×10.

Subarachnoid hemorrhages in the central nervous system.

(4/5). In the brain, gross and microscopic congestion was found(2/5, 40%).

The heart showed no gross lesions in this group, whereasacute cardiomyocyte changes, intracytoplasmic vacuolization,and hypereosinophilia were detected microscopically in 100%of animals (5/5). Mild congestion (2/5, 40%), mild hemorrhages(2/5, 40%) as well as mild interstitial edema (1/5, 20%) were alsoobserved in this group.

Hepatic congestion was present in four animals out of five(80%) in this group and mild congestion (5/5, 100%) withmild hepatocytic vacuolization (3/5, 60%). Renal congestionwas observed in 40% (2/5), while congestion was observed in100% (5/5), with intravascular bubble-like cavities in 60% of theanimals (3/5).

Other findings in this group were congestion of the adiposetissue (2/5, 40%) and muscle fiber hypereosinophilia as well aswavy fibers in two animals (2/5, 40%).

Comparative ResultsAs shown in Figure 1, where the different gas scoremodes of eachgroup are represented, both the control and the C/D euthanized

animals do not present macroscopic bubbles in any definedlocation for the gas score. The C/D mortality group presentsbubbles in all locations, varying between abundant number ofbubbles and completely filling vessel sections, except the presenceof few bubbles leading to interstitial emphysema and the absenceof subcapsular emphysema.

As shown in Tables 1, 2, the presence of microscopic bubble-like cavities was another finding also observed with greaterincidence in the group of C/D mortality, with these bubblesobserved in capillaries and small-sized blood vessels of the lung(38%), the heart (25%), and the kidney (25%). In the C/Deuthanized group, the presence of these microscopic bubble-like cavi ties was lower in kidney, although with a relevantpercentage (60%).

Pulmonary and splenic emphysema was observed in bothgroups. Emphysema in the adipose tissue was only seen inthe C/D mortality group. Hemorrhages were more prevalent indifferent organs of the C/D mortality group, with hemorrhagespresent in the lung (50%), heart (50%), and brain (25%). The C/Deuthanized group presented lower hemorrhages in the lung (20%)and heart (40%). No hemorrhages were seen in the brain.

Other relevant lesions observed were interfibrillar edema incardiac (38% in C/D mortality and 20% in C/D euthanized)and skeletal muscles (50% in C/D mortality group) as well ashypereosinophilia in these fibers (88% in C/D mortality groupand 40% in C/D euthanized group). Congestion was seen in mostof the organs in all groups.

DISCUSSION

The main pathological finding in rabbits that died byexperimental compression/decompression was the presenceof a large amount of gas bubbles widely distributed throughoutboth the central and peripheral venous circulation. Emphysema(mainly in the lung, spleen, and adipose tissue) and hemorrhagesin the lung, heart, and brain were the second main grossand histological finding in those rabbits. These pathologicalfindings have also been described in beaked whales (FamilyZiphiidae) and Risso’s dolphins (Grampus griseus) that died ofdecompression-like sickness (7–12) (Figure 3D). Nevertheless,these lesions were absent in the rabbits that survived for 1 hafter decompression.

The abundance, distribution, and gas composition of theanimals from this study have been previously described indetail by Bernaldo de Quirós et al. (21). Thus, we will onlybriefly summarize those results here in order to compare themwith the results from cetaceans. Macroscopic gas was observedmassively and systematically distributed in rabbits that dieddue to a compression/decompression protocol. These resultsare in agreement with other studies such as those of Eggletonet al. (14), Lever et al. (16), or Shim et al. (17), carried outwith guinea pigs, mice, and rabbits, respectively. In strandedcetaceans, the abundance and distribution of macroscopic gaswere also analyzed in the two stranded Risso’s dolphins diagnosedwith a decompression-like sickness as the cause of death, beingalso systemically distributed (12) (Figures 2D,F).

Frontiers in Veterinary Science | www.frontiersin.org 5 June 2021 | Volume 8 | Article 676499

Velázq

uez-W

allra

fetal.

Decompressive

PathologyThroughExp

erim

entalM

odel

TABLE 1 | Macroscopic findings in each group of rabbits and organs.

Lung Heart Thymus Liver Kidney Spleen Brain Fat

Congestion Edema Emphysema Pneumonia Hemorrhages Petechiae Congestion Congestion Congestion Emphysema Congestion Congestion Emphysema

C/DMortality 3/8 (38%) 2/8 (25%) 6/8 (75%) 1/8 (13%) 3/8(38%) 0/8(0%) 5/8 (63%) 4/8 (50%) 1/8 (13%) 3/8 (38%) 4/8 (50%) 4/8 (50%) 7/8 (88%)

C/DEuthanized 0/5 (0%) 1/5 (20%) 2/5 (40%) 0/5 (0%) 0/5 (0%) 0/5 (0%) 4/5 (80%) 2/5 (40%) 2/5 (40%) 1/5 (20%) 2/5 (40%) 2/5 (40%) 0/5 (0%)

Control 3/4 (75%) 0/4 (0%) 0/4 (0%) 1/4 (25%) 0/4 (0%) 3/4 (75%) 3/4 (75%) 3/4 (75%) 2/4 (50%) 0/4 (0%) 1/4 (25%) 0/4 (0%) 0/4 (0%)

TABLE 2 | Microscopic findings in each group of rabbits and organs.

Lung Heart

Bubble-like

cavities

Congestion Edema Emphysema Hemorrhages Pneumonia Thrombi Bubble-like

cavities

Congestion Contraction

band

necrosis

Edema Hemorrhages Hyper-

eosinophilia

Vacuolization

C/D

mortality

3/8 (38%) 7/8 (88%) 0/8 (0%) 8/8 (100%) 4/8 (50%) 1/8 (13%) 0/8 (0%) 2/8 (25%) 7/8 (88%) 1/8 (13%) 3/8 (38%) 4/8 (50%) 7/8 (88%) 7/8 (88%)

C/D

euthanized

0/5 (0%) 3/5 (60%) 0/5 (0%) 5/5 (100%) 1/5 (20%) 0/5 (0%) 0/5 (0%) 0/5 (0%) 2/5 (40%) 0/5 (0%) 1/5 (20%) 2/5 (40%) 5/5 (100%) 5/5 (100%)

Control 0/4 (0%) 3/4 (75%) 0/4 (0%) 0/4 (0%) 0/4 (0%) 1/4 (25%) 0/4 (0%) 0/4 (0%) 1/4 (25%) 0/4 (0%) 0/4 (0%) 0/4 (0%) 2/4 (50%) 2/4 (50%)

Liver Kidney Spleen Brain Skeletal muscle

Congestion Vacuolization Bubble-like

cavities

Congestion Congestion Emphysema Congestion Hemorrhages Edema Hypereosinophilia Wavy fibers

C/D

mortality

5/8 (63%) 4/8 (50%) 2/8 (25%) 8/8 (100%) 7/8 (88%) 6/8 (75%) 4/8 (50%) 2/8 (25%) 4/8 (50%) 7/8 (88%) 1/8 (13%)

C/D

euthanized

5/5 (100%) 3/5 (60%) 3/5 (60%) 5/5 (100%) 5/5 (100%) 4/5 (80%) 2/5 (40%) 0/5 (0%) 0/5 (0%) 2/5 (40%) 2/5 (40%)

Control 1/4 (25%) 1/4 (25%) 0/4 (0%) 3/4 (75%) 3/4 (75%) 0/4 (0%) 0/4 (0%) 0/4 (0%) 0/4 (0%) 0/4 (0%) 0/4 (0%)

Frontiers

inVeterin

ary

Science|w

ww.fro

ntiersin

.org

6Ju

ne2021|Volume8|A

rticle676499

Velázquez-Wallraf et al. Decompressive Pathology Through Experimental Model

Microscopic bubble-like cavities (i.e., small round to ovalnon-staining spaces that sometimes displaced erythrocytes)were observed within blood capillaries and small vessels fromthe lung, heart, and kidneys of the rabbits that died afterdecompression as well as in the kidney of animals that surviveddecompression and were euthanized. Microscopic gas embolismin the lung small pulmonary arteries, capillaries, and veinshas been previously described, such as that of Geng et al. (4)in decompressed rabbits. L’Abbate et al. (24) also describedmicroscopic bubbles in hepatic sinusoids from rats. In cetaceansdiagnosed with a decompression-like sickness as the causeof death, abundant microscopic gas embolism was observedin renal capillaries, subcapsular veins, hepatic sinusoids, andpulmonary (Figure 3B), coronary, intestinal, and meningealvessels (7, 10, 12).

These microscopic lesions were also observed in cetaceans,disrupting the white matter structure of the brain and spinalcord (7, 8). Similarly, micro-bubbles in the nervous system havebeen described in human medicine, being primarily seen in thespinal cord (25). It has been hypothesized that the low vascularsupply and the high lipid content of the spinal white matter,conferred by the myelin that covers the axons, increase theaffinity of inert gases for this structure (26). Thus, most CNSlesions are described in the spinal cord’s white matter, such aspunctured hemorrhages, spongiosis, axon swelling, and myelindegeneration (27). Microscopic bubble-like cavities were notobserved in the brain or cranial spinal cord of the rabbits, butonly the cranial part of the spinal cord was sampled. Futurestudies should aim at investigating the entire spinal cord.

Gas composition analysis of the gas embolism was performedin the rabbits of this study (21) and in some of the cetaceansdiagnosed with a decompression-like sickness. These includeda beaked whale stranded in association with naval exercises(11) and Risso’s dolphins after a deadly prey interaction (12).In all cases, nitrogen was the main compound, followed byCO2. Hydrogen, a putrefaction marker, was absent or present inlow quantities.

The most affected organ in all the rabbits from the C/Dmodel was the lung. Mild to severe pulmonary emphysemawas observed in all of them, while no control rabbits showedpulmonary emphysema. Similar results have been observed inother animal models of decompressive sickness with rats andrabbits (4, 19, 20, 28) as well as in mass stranded beaked whalesassociated with military exercises and in single stranded Risso’sdolphins analyzed in this comparative study (7, 12) (Figure 2B).

Emphysema in other locations, such as the spleen and theadipose tissue of the abdominal cavity and mesenteric areas,was also observed in rabbits. Although splenic emphysemawas observed to be affecting both groups in the C/D model,the severity was more critical in the rabbits that died afterdecompression, with severe emphysematous spleens vs. the mildemphysema of the spleens of rabbits that survived decompressionand were euthanized. Clay (15) also described that half of thedogs analyzed presented macroscopic and microscopic gas in thespleen, which sometimes displaced the splenic follicles.

The adipose tissue (i.e., mesenteric, abdominal, and coronaryfat depots) presented mild multifocal emphysema in most rabbits

that died after decompression. Since nitrogen is more soluble infatty tissues than in non-fat tissues (17), the relevant presence ofbubbles within the adipose tissue in animals that have died byDCS was probably a macroscopic finding to be considered in theassessment of this disease. In the case of the cetaceans diagnosedwith DCS, emphysema in the adipose tissue was evident in mostcases, being more evident in the coronary fatty deposits andbeneath the renal capsule (7, 12).

Another relevant injury found in the rabbits that died afterdecompression was hemorrhages in different organs. Bubblescan cause vasoconstriction, leading to the presence of ischemia,edema, and hemorrhages in target organs such as the lung (3).Severe pulmonary hemorrhages were present in the rabbits thatdied from compression–decompression than those that surviveddecompression after 1 h. Pulmonary hemorrhages have also beendescribed in rats (19, 29), rabbits (4), and stranded cetaceans withpathological signs of DCS (7, 12).

The beaked whales in the mass strandings were alldiagnosed with decompressive-like sickness (7, 8) that showedmacroscopically acute disseminated hemorrhages in differentorgans, being especially severe in the CNS. These multifocalhemorrhages were mainly in subarachnoid areas, spinal cord,and meninges (Figure 3F). These findings are similar to thosepresented in rabbits from this study. In addition to hemorrhagesin the CNS, vascular congestion, myelin degeneration, axonswelling, and pericapillary edema are common findings in pigs,humans, or rats (19, 26, 27, 30). However, in this experimentalmodel, only congestion and brain-associated hemorrhages in thesubarachnoid space were observed.

Interstitial and alveolar pulmonary edema was observed inall groups. This edema has also been macroscopically describedin other experimental models that reproduced DCS in rats,sheep, and rabbits (4, 20, 28, 31). In these models, pulmonaryedema was one of the most observed lesions, along withemphysema. According to Atkins et al. (31), pulmonary edemais related to the development of pulmonary hypertension andincreased permeability of blood capillaries due to the contactof microbubbles with the endothelium, inducing the releaseof intracellular calcium, causing damage to endothelial cells,increasing their permeability, and allowing the release of protein-rich fluid into the intracellular space (20). In this experimentalmodel, the low number of affected rabbits and all groups’presence do not seem relevant to this finding. Cetaceansdiagnosed with DCS also had diffuse pulmonary edema (8) andnon-specific lesions linked to different causes of death.

Acute muscle changes were associated with ischemic damagecaused by stressful situations. In this study, these changeswere found in muscular tissues such as skeletal muscle andmyocardium. These acute changes usually occur within minutesafter ischemia, including contraction band necrosis and wavyfibers’ presence. These changes are well studied in other recentstudies that analyzed the stress to which cetaceans were exposedwhile stranding alive (32) (Figure 3C). In this study, thesetwo lesions were reported in fewer animals than expected (1/8animals in the mortality group presented both lesions and 2/5animals in the euthanized group presented wavy fibers). Otheracute changes such as hypereosinophilia and intracytoplasmic

Frontiers in Veterinary Science | www.frontiersin.org 7 June 2021 | Volume 8 | Article 676499

Velázquez-Wallraf et al. Decompressive Pathology Through Experimental Model

vacuolization were observed in the cardiac and skeletal muscleof animals that died after decompression and, to a lesser extent,in those that survived and were euthanized. Since in the rabbitsall the procedures were carried out under surgical anesthesia, thismight prevent the appearance of some stress-related lesions.

Vacuolization of hepatocytes was observed in the C/D model,with the rabbits that died by the protocol being more affected.L’Abbate et al. (24) conducted a study on the changes observed inrats’ liver after undergoing a rapid decompression protocol. Thus,hepatocellular vacuolization was not observed in spontaneousdeath or in the group euthanized after 3 h, but it was observedin the animals euthanized at 24 h, with different severity levels.These findings are dissimilar to those obtained in our study,where the animals that died shortly after decompression hadmore marked hepatocellular vacuolization than those that wereeuthanized at 1 h post-decompression.

While in this study no fibrin microthrombi were observedin the compression/decompression model, pulmonary arterialmicrothrombi have been described in other studies of DCSwith similar protocols, such as in the study of Tanoue et al.(33), where rabbits were exposed to a compression protocolof 6 ATA for 40min and rapid decompression of 5min,and the animals euthanized immediately after decompressionshowed these microthrombi in large arteries of the lung, orin the study of Geng et al. (4), where thrombosis was seenin small pulmonary arteries, capillaries, and veins (7’98 ATAfor 1 h, rapid decompression for 5min in rabbits). Arieliet al. (18), with a rat model subjected to 12’49 ATA for33min and a rapid decompression in 6min, also describedthe blood alterations generated by the microbubbles and theplatelets’ consequent activation which increased the presence ofmicrothrombi and disseminated intravascular coagulation. Incetaceans, the presence of these microthrombi associated withdecompression-like sickness was not observed.

In summary, it is necessary to highlight the difference betweenthe severe presence of systemic gas embolism and associated gaslesions in rabbits dead by decompression vs. the absence or lowerincidence in euthanized animals. Other studies have observedthat, despite exposing individuals with a similar profile (species,sex, age, and weight) to the same protocol, bubble formation andlethality are highly variable (17, 24). Based on this, cetaceansexposed to the same diving profile and subjected to the samestress can present different results, with some animals developinga lethal DCS, while others may survive.

In conclusion, the rabbits that died after decompressionpresented large quantities of macroscopic and microscopic gasbubbles systemically distributed, emphysema, and hemorrhagesin multiple vital organs. Most of the lesions described wereprobably due to the bubbles’ mechanical and embolic damage.

These same lesions have been described in cetaceans, consistentwith a decompression-like sickness, reinforcing the pathologicalfindings found. Besides this, almost half of the rabbits thatsurvived for 1 h after decompression did not show the samelesions or severity. It reveals that individuality plays an essentialrole in this disease as it has previously been hypothesized inanimal models and human diving medicine.

DATA AVAILABILITY STATEMENT

The original contributions presented in the study are includedin the article/supplementary material, further inquiries can bedirected to the corresponding authors.

ETHICS STATEMENT

The animal study was reviewed and approved by NorwegianCommittee for Animal Experiments (2154) and the EthicalCommittee for Animal Experiments of the University of lasPalmas de Gran Canaria (CEEBA-HUGCDN 002/2010).

AUTHOR CONTRIBUTIONS

AF, AM, and YB took charge of conceptualization. AF and AMtook charge of funding. AV-W contributed to writing. YB, MC,AM, MA, and AV-W contributed to the experimental proceduresand laboratory analyses. AV-W, YB, MC, and MA took chargeof the pathological studies. All the authors contributed to reviewand editing. AF, YB, and MC supervised the study.

FUNDING

This study was supported by the National Project PGC2018-101226-B-I00 and the Canary Islands Government, which hasfunded and provided support to the stranding network. AV-Wwas funded by the University Professor Formation fellowshipfrom the Spanish Ministry of Education (FPU17/00763). Thiswork was also supported by the Liaison Committee between theCentral Norway Regional Health Authority and the NorwegianUniversity of Science and Technology, grant number 46028600.

ACKNOWLEDGMENTS

The authors would like to thank all colleagues from theUniversity of las Palmas de Gran Canaria (Spain) andvolunteers of the Cetacean Stranding Network who contributedto this work and the hyperbaric medicine division of theNorwegian University of Science and Technology (Norway) forits scientific contribution.

REFERENCES

1. Vann RD, Butler FK, Mitchell SJ, Moon RE. Decompression illness. Lancet.(2011) 377:153–64. doi: 10.1016/S0140-6736(10)61085-9

2. Edmonds C, Bennett M, Lippmann J, Mitchell SJ.Diving Subaquatic Medicine.5th ed. Boca Raton, FL: CRC Press (2015). doi: 10.1201/b18700

3. Pekka S, Knight B. Knight’s Forensic Pathology. 3rd ed. London: EdwardArnold Ltd (2004). p. 488–90.

Frontiers in Veterinary Science | www.frontiersin.org 8 June 2021 | Volume 8 | Article 676499

Velázquez-Wallraf et al. Decompressive Pathology Through Experimental Model

4. Geng M, Zhou L, Liu X, Li P. Hyperbaric oxygen treatment reducedthe lung injury of type II decompression sickness. Int J Clin Exp Pathol.

(2015) 8:1791–803.5. Ponganis PJ, Koyman GL, Ridway SH. Comparative diving physiology. In:

Brubakk AO, Neuman TS, editors. Bennett and Elliott’s Physiology and

Medicine of Diving. 5th ed. New York, NY: Saunders Elsevier Science Ltd(2003). p. 221–6.

6. Jepson PD, ArbeloM, Deaville R, Patterson I, Castro P, Baker, et al. Gas-bubblelesions in stranded cetaceans.Nature. (2003) 425:575–6. doi: 10.1038/425575a

7. Fernández A, Edwards J, Rodríguez F, de los Monteros A, Herráez P, Castro,et al. “Gas and fat embolic syndrome” involving a mass stranding of beakedwhales (Family Ziphiidae) exposed to anthropogenic sonar signals.Vet Pathol.(2005) 42:446–57. doi: 10.1354/vp.42-4-446

8. Fernández A, Sierra E, Martín V,MéndezM, Sacchini S, Bernaldo de Quirós Y,et al. Last “Atypical” beaked whales mass stranding in the canary islands (July,2004). J Marine Sci Res Dev. (2012) 2:2. doi: 10.4172/2155-9910.1000107

9. Bernaldo de Quirós Y, Fernández A, Baird RW, Brownell RL, Jr, Aguilardel Soto N, Allen D, et al. Advances in research on the impacts of anti-submarine sonar on beaked whales. Proc R Soc B. (2019) 286:20182533.doi: 10.1098/rspb.2018.2533

10. Arbelo M, Bernaldo de Quirós Y, Sierra E, Méndez M, Godinho A,Ramírez G, et al. Atypical beaked whale mass stranding in Almeria’scoast: pathological study. Int J Anim Sound Rec. (2008) 17:293–323.doi: 10.1080/09524622.2008.9753853

11. Bernaldo de Quirós Y, González-Díaz Ó, Saavedra P, Arbelo M, SierraE, Sacchini, et al. Methodology for in situ gas sampling, transport andlaboratory analysis of gases from stranded cetaceans. Sci Report. (2011) 1:193.doi: 10.1038/srep00193

12. Fernández A, Sierra E, Díaz-Delgado J, Sacchini S, Sanchez-Paz Y, Suarez-Santana C, et al. Deadly acute decompression sickness in risso’s dolphins. SciRep. (2017) 7:13621. doi: 10.1038/s41598-017-14038-z

13. García-Párraga D, Crespo-Picazo JL, Bernaldo de Quirós Y, Cervera V, Martí-Bonmati L, Díaz-Delgado J, et al. Decompression sickness (‘the bends’) in seaturtles. Dis Aquat Org. (2014) 111:191–205. doi: 10.3354/dao02790

14. Eggleton P, Elsden S, Fegler J, Hebb C. A study of the effects ofrapid ‘decompression’ in certain animals. J Physiol. (1945) 104:129–50.doi: 10.1113/jphysiol.1945.sp004111

15. Clay JR. Histopathology of experimental decompression sickness. AerospMed. (1963) 34:1107–10.

16. Lever MJ, Miller KW, Paton WDM, Smith EB. Experiments on the genesisof bubbles as a result of rapid decompression. J Physiol. (1966) 184:964–9.doi: 10.1113/jphysiol.1966.sp007960

17. Shim SS, Patterson FP, Kendall MJ. Hyperbaric chamber and decompressionsickness: an experimental study. Can Med Assoc J. (1967) 97:1263–72.

18. Arieli R, Boaron E, Abramovich A. Combined effect of denucleation anddenitrogenation on the risk of decompression sickness in rats. J Appl Physiol.(2009) 106:1453–8. doi: 10.1152/japplphysiol.91146.2008

19. Ni X, Cai Z, Fan D, Liu Y, Zhang R, Liu, et al. Protective effect of hydrogen-rich saline on decompression sickness in rats.Aviat Space EnvironMed. (2011)82:604–9. doi: 10.3357/ASEM.2964.2011

20. Tang S, Liao W, Wu S, Pao H, Huang K, Chu S. The blockade of store-operated calcium channels improves decompression sickness in rats. FrontPhysiol. (2020) 10:1616. doi: 10.3389/fphys.2019.01616

21. Bernaldo de Quirós Y, González-Díaz O, Møllerløkken A, BrubakkA, Hjelde A, Saavedra, et al. Differentiation at autopsy between invivo gas embolism and putrefaction using gas composition analysis.Int J Legal Med. (2013) 127:437–45. doi: 10.1007/s00414-012-0783-6

22. Bernaldo de Quirós Y, Saavedra P, Møllerløkken A, Brubakk A, JørgensenA, González-Díaz, et al. Differentiation at necropsy between in vivo gasembolism and putrefaction using a gas score. Vet Sci Res J. (2016) 106:48–55.doi: 10.1016/j.rvsc.2016.03.007

23. Suvarna S, Layton C, Bancroft J. Bancroft’s Theory and Practice of HistologicalTechniques. 7th ed. Philadelphia: Churchill Livingstone of Elsevier (2018).

24. L’Abbate A, Kusmic C, Matteucci M, Pelosi G, Navari A, Pagliazzo, et al.Gas embolization of the liver in a rat model of rapid decompression.Am J Physiol Regul Integr Comp Physiol. (2010) 299:R673–82.doi: 10.1152/ajpregu.00699.2009

25. Caruso JL. Pathology of diving accidents. In: Brubakk AO, Neuman TS,editors. Bennett Elliott’s: Physiology Medicine of Diving. 5th ed. Eastbourne:Saunders Elsevier (2003). p. 729–43.

26. Francis T, Mitchell S. Pathophysiology of decompression sickness. In: Bove A,editor. Bove Davis’ Diving Medicine. 4th ed. Philadelphia: Saunders (2004). p.165–83. doi: 10.1016/B978-0-7216-9424-5.50014-9

27. Broome JR, Dick EJ, Jr. Neurological decompression illness in swine. AviatSpace Environ Med. (1996) 67:207–13.

28. Zhang K, Wang D, Jiang Z, Ning X, Buzzacott P, Xu W. Endothelialdysfunction correlates with decompression bubbles in rats. Sci Rep. (2016)6:33390. doi: 10.1038/srep33390

29. Bao X, Chen H, Fang Y, Yuan H, You P, Ma J, et al. Clopidogrel reducesthe inflammatory response of lung in a rat model of decompression sickness.Respir Physiol Neurobiol. (2015) 211:9–16. doi: 10.1016/j.resp.2015.02.003

30. Dick EJ, Jr, Broome JR, Hayward IJ. Acute neurologic decompression illnessin pigs: lesions of the spinal cord and brain. Lab Anim Sci. (1997) 47:50–7.

31. Atkins C, Lehner C, Beck K, Dubielzig R, Nordheim E, Lanphier E.Experimental respiratory decompression sickness in sheep. J Appl Physiol.(1988) 65:1163–71. doi: 10.1152/jappl.1988.65.3.1163

32. Câmara N, Sierra E, Fernández-Maldonado C, Espinosa de los Monteros A,Arbelo M, Fernández A, et al. Stress cardiomyopathy in stranded cetaceans: ahistological, histochemical and immunohistochemical study. Vet Rec. (2019)185:694. doi: 10.1136/vr.105562

33. Tanoue K, Mano Y, Kuroiwa K, Suzuki H, Shibayama M, Yamazaki H.Consumption of platelets in decompression sickness of rabbits. J Appl Physiol.(1987) 62:1772–9. doi: 10.1152/jappl.1987.62.5.1772

Conflict of Interest: The authors declare that the research was conducted in theabsence of any commercial or financial relationships that could be construed as apotential conflict of interest.

Copyright © 2021 Velázquez-Wallraf, Fernández, Caballero, Møllerløkken, Jepson,

Andrada and Bernaldo de Quirós. This is an open-access article distributed under the

terms of the Creative Commons Attribution License (CC BY). The use, distribution

or reproduction in other forums is permitted, provided the original author(s) and

the copyright owner(s) are credited and that the original publication in this journal

is cited, in accordance with accepted academic practice. No use, distribution or

reproduction is permitted which does not comply with these terms.

Frontiers in Veterinary Science | www.frontiersin.org 9 June 2021 | Volume 8 | Article 676499